Keyword Research Ninja Tactics – SMX London 2011
I always wanted to be a ninja… Not a ninja of links though. Which is why it’s awesome my vocational change has brought me back into the realm of them. SEO ninja’s are unlike common ninja’s and drink coffee.
This is the second group of presentations I attended on day 2 of SMX London breaks down keyword research, to a degree which most of you should find very interesting. Watch for the tips, they are worth a takeaway or two.
Richard Baxter – SEO Gadget
How does anyone make the data they get actionable? How does one get from a data set to a design and more importantly into SEO. The key with research is finding a way of grouping the findings, creating plans and actioning it.
You need to look at long tail keywords to figure out how users are searching for other products. By grouping and identifying segments of a long tail keyword you can quickly create groups and segment again based on these. An example is:
Black Audi A3 Greater London
Most of our most important traffic comes from the longest tail traffic 70th percentile.
- Datasets allow you to filter and categorise keywords
- Allow you to work out and create an order for how people search
- Allows you to make decisions about how your pages and on-page information is categorised.
Richard goes through awesome excel set of excel queries:
Man I love excel, and am always so impressed by the stuff that comes out of SEO gadget. It seems array formulas are the key for dissecting a number of queries. There is definitely a skill to learning excel in a presentations, a great number of people are scrambling around trying to get those cogs turning.
So Richard goes on to explain that once you have your methodology down, it’s time to increase the volume of data – more data = better decisions. Agree, agree, and agree.
Deep dive through the Google suggest API and turn 1000 keyword into 10000, categorise and segment these and drill down again. It could help you find an ‘uncategorised’ group which you aren’t even targeting!!!
All of Richards tools are listed here, enjoy – http://bit.ly/jai89j
Christine Churchill – KeyRelevance
Always love an overload of macros… which is what just happened here. For all you tool lovers Christine’s presentation is going to get you excited.
We all know about the main SEO tools in the market but I hope that Christine shouts about some new ones. Christine explains that last September’s Google keyword tool change is because Google stopped pulling the data from its partner sites… sneaky.
We all love [exact] match and Christine does too. Christine also suggests using the advanced filtering tool with special focus being on the mobile devices – worth considering if you are optimising in a segment for which mobile searches are prominent.
Christine also likes Google Insights and Keyword discovery, which I whole heartedly agree with. I think a big take away at this stage in the presentation is to never get too comfortable with the toolset you use. New tools are out there, the functionality of which often solve the problems which appear in the gaps left by your existing set. Other tool info worth a mention is:
- YouTube suggest – untapped predictive text tool. Worth an extra special mention for those of you optimising video.
- Google Wonder Wheel – a tool which I personally haven’t used but will definitely worth a look.
- Microsoft’s ad centre – pulls live data… live is better than dead.
Lasse Clarke Storgaard – MediaCom
Searchonomics, a combination of my job and favourite A-level class, a great start! Plus Lasse looks like he fell out of a very cool Indy band, hopefully there is song.
He suggests slowing down; nobody cares about how quickly you do it. It’s all about the keywords that work in the end. He explains that he thinks of consumers as having 3 mindsets; one shelf, on which they look high and low. They are not looking for your product:
- Want to buy – 8%
- Consider – 10-15%
- Research 60-80% – not going to buy, but just want to know about it all.
The main concept is about creating a funnel and massaging a consumer down it – everyone loves a massage.
PPPI – Pay Per Purchase Intent
- Which is a set of concentric marketing segmentation circles around an onion. Making me hungry… but on with the ‘search philosophy’.
I think this presentation again highlights how much general CRM and marketing strategies are finding their way into ‘core’ search presentations. Which again highlight’s how there is an underlying shift towards focusing on content and the consumer… Ponder…
Another segmentation slide from Lasse. I read a text book about this in 2001; it was a good read though.
Kevin Gibbons – SEOptimise
Most searches are long tail, 94% of them in fact, I love long tail. There is more to love about these of a longer tail – they convert better and there is less bounce rate.
Idea – focus on the long tail, focus on it again, and then focus more.
Focusing on the long tail needs to be applied from an entire website perspective; you need to totally submerge the site with a long tail mentality.
For content ideas utilise Google Instant, dig deep into analytics and spot trends within – which can then be applied to KW research. Kev suggests creating bespoke segmentation throughout analytics (removing brand is a must) and creating a PPC campaign solely for keyword research something which we wholeheartedly agree with.
Top tip – Use impression share to figure out full search volume!
Top tip – Wikipedia traffic tool – USE IT
Hitwise – another tool which will help you figure out where to target. One which supports a big budget, but the general consensus is; you get back what you pay and then some in data.
It’s worth using as many references and tools as possible to create an average for metrics such as CTR & bounce rates. We recommend this for ecommerce sites in regards to predicting changes in demand and sales.
Kev also suggests dropping that data into excel and making predictions based on traffic volumes. Why not go one step further and apply CPC AdWords data and try to combine conversion data to make the info even more valuable.
Another take away from Kev – go long tail, but don’t go too far. How far is too far? Turning yourself into a fox!
Top takeaways from the Q&A session:
- Adwords API is a lot more granular than other Google data tools. Be very careful with GKT as it pulls in last month which could be very seasonal.
- Defensive ideas – look to the future, get an idea and make sure you get it up ASAP. Logical really, dragging your heels never helps anyone.
- NSFG – Scraping Google – distribute load via multiple IP’s I guess via a proxy of some kind.
- NSFG – Break load API tool – means you never set alarm bells ringing.